Ingo recently introduced a great speedup for allocating new mmaps using the
free_area_cache pointer which boosts the specweb SSL benchmark by 4-5% and
causes huge performance increases in thread creation.
The downside of this patch is that it does lead to fragmentation in the
mmap-ed areas (visible via /proc/self/maps), such that some applications
that work fine under 2.4 kernels quickly run out of memory on any 2.6
kernel.
The problem is twofold:
1) the free_area_cache is used to continue a search for memory where
the last search ended. Before the change new areas were always
searched from the base address on.
So now new small areas are cluttering holes of all sizes
throughout the whole mmap-able region whereas before small holes
tended to close holes near the base leaving holes far from the base
large and available for larger requests.
2) the free_area_cache also is set to the location of the last
munmap-ed area so in scenarios where we allocate e.g. five regions of
1K each, then free regions 4 2 3 in this order the next request for 1K
will be placed in the position of the old region 3, whereas before we
appended it to the still active region 1, placing it at the location
of the old region 2. Before we had 1 free region of 2K, now we only
get two free regions of 1K -> fragmentation.
The patch addresses thes issues by introducing yet another cache descriptor
cached_hole_size that contains the largest known hole size below the
current free_area_cache. If a new request comes in the size is compared
against the cached_hole_size and if the request can be filled with a hole
below free_area_cache the search is started from the base instead.
The results look promising: Whereas 2.6.12-rc4 fragments quickly and my
(earlier posted) leakme.c test program terminates after 50000+ iterations
with 96 distinct and fragmented maps in /proc/self/maps it performs nicely
(as expected) with thread creation, Ingo's test_str02 with 20000 threads
requires 0.7s system time.
Taking out Ingo's patch (un-patch available per request) by basically
deleting all mentions of free_area_cache from the kernel and starting the
search for new memory always at the respective bases we observe: leakme
terminates successfully with 11 distinctive hardly fragmented areas in
/proc/self/maps but thread creating is gringdingly slow: 30+s(!) system
time for Ingo's test_str02 with 20000 threads.
Now - drumroll ;-) the appended patch works fine with leakme: it ends with
only 7 distinct areas in /proc/self/maps and also thread creation seems
sufficiently fast with 0.71s for 20000 threads.
Signed-off-by: Wolfgang Wander <wwc@rentec.com>
Credit-to: "Richard Purdie" <rpurdie@rpsys.net>
Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Acked-by: Ingo Molnar <mingo@elte.hu> (partly)
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
A lot of the code in arch/*/mm/hugetlbpage.c is quite similar. This patch
attempts to consolidate a lot of the code across the arch's, putting the
combined version in mm/hugetlb.c. There are a couple of uglyish hacks in
order to covert all the hugepage archs, but the result is a very large
reduction in the total amount of code. It also means things like hugepage
lazy allocation could be implemented in one place, instead of six.
Tested, at least a little, on ppc64, i386 and x86_64.
Notes:
- this patch changes the meaning of set_huge_pte() to be more
analagous to set_pte()
- does SH4 need s special huge_ptep_get_and_clear()??
Acked-by: William Lee Irwin <wli@holomorphy.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This is the core of the (much simplified) early reclaim. The goal of this
patch is to reclaim some easily-freed pages from a zone before falling back
onto another zone.
One of the major uses of this is NUMA machines. With the default allocator
behavior the allocator would look for memory in another zone, which might be
off-node, before trying to reclaim from the current zone.
This adds a zone tuneable to enable early zone reclaim. It is selected on a
per-zone basis and is turned on/off via syscall.
Adding some extra throttling on the reclaim was also required (patch
4/4). Without the machine would grind to a crawl when doing a "make -j"
kernel build. Even with this patch the System Time is higher on
average, but it seems tolerable. Here are some numbers for kernbench
runs on a 2-node, 4cpu, 8Gig RAM Altix in the "make -j" run:
wall user sys %cpu ctx sw. sleeps
---- ---- --- ---- ------ ------
No patch 1009 1384 847 258 298170 504402
w/patch, no reclaim 880 1376 667 288 254064 396745
w/patch & reclaim 1079 1385 926 252 291625 548873
These numbers are the average of 2 runs of 3 "make -j" runs done right
after system boot. Run-to-run variability for "make -j" is huge, so
these numbers aren't terribly useful except to seee that with reclaim
the benchmark still finishes in a reasonable amount of time.
I also looked at the NUMA hit/miss stats for the "make -j" runs and the
reclaim doesn't make any difference when the machine is thrashing away.
Doing a "make -j8" on a single node that is filled with page cache pages
takes 700 seconds with reclaim turned on and 735 seconds without reclaim
(due to remote memory accesses).
The simple zone_reclaim syscall program is at
http://www.bork.org/~mort/sgi/zone_reclaim.c
Signed-off-by: Martin Hicks <mort@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch implements a number of smp_processor_id() cleanup ideas that
Arjan van de Ven and I came up with.
The previous __smp_processor_id/_smp_processor_id/smp_processor_id API
spaghetti was hard to follow both on the implementational and on the
usage side.
Some of the complexity arose from picking wrong names, some of the
complexity comes from the fact that not all architectures defined
__smp_processor_id.
In the new code, there are two externally visible symbols:
- smp_processor_id(): debug variant.
- raw_smp_processor_id(): nondebug variant. Replaces all existing
uses of _smp_processor_id() and __smp_processor_id(). Defined
by every SMP architecture in include/asm-*/smp.h.
There is one new internal symbol, dependent on DEBUG_PREEMPT:
- debug_smp_processor_id(): internal debug variant, mapped to
smp_processor_id().
Also, i moved debug_smp_processor_id() from lib/kernel_lock.c into a new
lib/smp_processor_id.c file. All related comments got updated and/or
clarified.
I have build/boot tested the following 8 .config combinations on x86:
{SMP,UP} x {PREEMPT,!PREEMPT} x {DEBUG_PREEMPT,!DEBUG_PREEMPT}
I have also build/boot tested x64 on UP/PREEMPT/DEBUG_PREEMPT. (Other
architectures are untested, but should work just fine.)
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Appended patch will setup compatibility mode TASK_SIZE properly. This will
fix atleast three known bugs that can be encountered while running
compatibility mode apps.
a) A malicious 32bit app can have an elf section at 0xffffe000. During
exec of this app, we will have a memory leak as insert_vm_struct() is
not checking for return value in syscall32_setup_pages() and thus not
freeing the vma allocated for the vsyscall page. And instead of exec
failing (as it has addresses > TASK_SIZE), we were allowing it to
succeed previously.
b) With a 32bit app, hugetlb_get_unmapped_area/arch_get_unmapped_area
may return addresses beyond 32bits, ultimately causing corruption
because of wrap-around and resulting in SEGFAULT, instead of returning
ENOMEM.
c) 32bit app doing this below mmap will now fail.
mmap((void *)(0xFFFFE000UL), 0x10000UL, PROT_READ|PROT_WRITE,
MAP_FIXED|MAP_PRIVATE|MAP_ANON, 0, 0);
Signed-off-by: Zou Nan hai <nanhai.zou@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Andi Kleen <ak@muc.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch fixes handling of accesses to ar.rsc via ptrace & restore_sigcontext
[With Thanks to Chris Wright for noticing the restore_sigcontext path]
Signed-off-by: Matthew Chapman <matthewc@hp.com>
Acked-by: David Mosberger <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Remove "pci=routeirq" option for ia64. This was a workaround
after ACPI IRQ routing was changed from "all at boot for everything
in _PRT" to "do it when the device is enabled" in case there were
drivers that didn't use pci_enable_device().
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
The nested_dtlb_miss handler currently does not handle fault from
hugetlb address correctly. It walks the page table assuming PAGE_SIZE.
Thus when taking a fault triggered from hugetlb address, it would not
calculate the pgd/pmd/pte address correctly and thus result an incorrect
invocation of ia64_do_page_fault(). In there, kernel will signal SIGBUS
and application dies (The faulting address is perfectly legal and we
have a valid pte for the corresponding user hugetlb address as well).
This patch fix the described kernel bug. Since nested_dtlb_miss is a
rare event and a slow path anyway, I'm making the change without #ifdef
CONFIG_HUGETLB_PAGE for code readability. Tony, please apply.
Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
printk() calls should include appropriate KERN_* constant.
Signed-off-by: Christophe Lucas <clucas@rotomalug.org>
Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Allow the SGI simulator (medusa) to work on generic kernels. There is
no inherent dependency on an sn2-specific kernel.
Boot tested on Altix, medusa and HP rx2600.
Signed-off-by: Greg Edwards <edwardsg@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Refresh arch/ia64/defconfig, as it was getting a bit stale. The only
manual changes I made were:
CONFIG_SCSI_SATA=y needed for some Altix base I/O cards
CONFIG_SCSI_SATA_VITESSE=y
CONFIG_DM_MULTIPATH=m the rest are already modules
CONFIG_FUSION_SPI=y new driver breakout
CONFIG_FUSION_FC=m
CONFIG_SGI_TIOCX=y enable some other SGI drivers
CONFIG_SGI_MBCS=m
CONFIG_AGP_SGI_TIOCA=m
Boot tested on Altix, HP rx2600 and Intel Tiger
Signed-off-by: Greg Edwards <edwardsg@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Patch from Lennert Buytenhek
The ixp2000 defconfigs are among the few that do not enable module
support by default. I keep enabling module support by hand for every
new kernel version, so let's just make this change upstream.
Signed-off-by: Lennert Buytenhek
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Richard Purdie
Fix typo in sharpsl_param.c so it works correctly on collie.
Signed-off-by: Richard Purdie
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Lennert Buytenhek
The IXP2000 has four timers, but if we're on an A-step IXP2800, timer
2 and 3 don't work. We need two timers for timekeeping (one for the
timer interrupt and one for tracking missed jiffies), so on early
IXP2800s we have no other choice but to use timer 1 and 4 for that,
but on all other IXP2000s we'd rather leave timer 4 free since that's
the only timer we can use for the watchdog.
So, on buggy IXP2000s (i.e. the A-step IXP2800) we use timer 4 for
tracking missed jiffies, and on all all non-buggy IXP2000s (i.e.
everything but the A-step IXP2800) we use timer 2.
On a pre-production IXP2800, this patch should print these messages
on boot:
Enabling IXP2800 erratum #25 workaround
Unable to use IXP2000 watchdog due to IXP2800 erratum #25
On any non-buggy IXP2800 (as well as on IXP2400s) you shouldn't see
anything at all, and the watchdog should be usable again.
Signed-off-by: Lennert Buytenhek
Signed-off-by: Deepak Saxena
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Catalin Marinas
This patch adds PCI support for the Versatile PB926 platform.
Signed-off-by: Colin King
Signed-off-by: Catalin Marinas
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Bellido Nicolas
Core support for AAEC-2000 based platforms.
This is an updated version of the previous patch, and takes
into account Russell's comments.
AAED-2000 default configuration will follow as soon
as some problems with the bootloader are sorted out...
Signed-off-by: Nicolas Bellido
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
The latest assembler catches this typo. (reported by Jim Wilson).
Signed-off-by: David Mosberger-Tang <davidm@hpl.hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
A fairly recent platform requirement states that the OS must clear the
whole TCE table at setup time, in case firmware left any active
mappings in it. Without this initialization, dynamic bus removes can
fail. Firmware rejects these requests if active mappings still exist
for a slot that has been deallocated by the OS.
Signed-off-by: John Rose <johnrose@austin.ibm.com>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Use the new cpu_has_feature macros instead of open coding it.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
Some rtasd printks were too loud. They would appear on a quiet boot
even though they were only informational.
Signed-off-by: Anton Blanchard <anton@samba.org>
Signed-off-by: Paul Mackerras <paulus@samba.org>
When DMA bounce buffers were unmapped and the data was memcpy'd to
the original buffer, we were not ensuring that the data was written
to RAM. This means that there was the potential for page cache
pages to have different cache states depending whether they've been
bounced or not.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
flush_dcache_page() did nothing for these caches, but since they
suffer from I/D cache coherency issues, we need to ensure that data
is written back to RAM.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
We need to re-initialise the stack pointers for undefined, IRQ
and abort mode handlers whenever we resume.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Create a temporary page table to startup secondary processors. This
page table must have a 1:1 virtual/physical mapping for the kernel
in addition to the standard mappings to ensure that the secondary
CPU can enable its MMU safely.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Nicolas Pitre
Without this some devices fail to work again after a suspend event.
Signed-off-by: Nicolas Pitre
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Catalin Marinas
The GPIO base for Integrator/CP is different from the
Integrator/AP. This patch sets the correct value for
INTEGRATOR_GPIO_BASE.
Signed-off-by: Catalin Marinas
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Patch from Catalin Marinas
The current red and blue colours on the Versatile CLCD are
reversed when the 5:6:5 mode is used. The patch sets the proper
bit in the SYS_CLCD register value.
Signed-off-by: Catalin Marinas
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
enable cpusets
enable new lpfc and jsm drivers
enable new dm-multipath
leave new agp disabled
disable rivafb, it does not handle the cards in G5 models (FX5200 as example)
the new nvidiafb doesnt work on bigendian, yet
Signed-off-by: Olaf Hering <olh@suse.de>
Acked-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Here is a patch to update the example configs in arch/ppc64/configs.
Signed-off-by: Paul Mackerras <paulus@samba.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch causes the ignore_normal_resume flag to be set slightly earlier,
before there is a chance that the apm driver will receive the normal resume
event from the BIOS. (Addresses Debian bug #310865)
Signed-off-by: Thomas Hood <jdthood@yahoo.co.uk>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch merges a lot of duplicated code in the slip and slirp drivers,
abstracts out the slip protocol, and makes the slip driver work in 2.6.
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Convert the boot-time host ptrace testing from clone to fork. They were
essentially doing fork anyway. This cleans up the code a bit, and makes
valgrind a bit happier about grinding it.
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Fix a build failure when CONFIG_MODE_SKAS is disabled and make a Makefile
comment fit in 80 columns.
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
A few files include the same header twice.
Signed-off-by: Jeff Dike <jdike@addtoit.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Wolfgang Wander